"... In PAGE 11: ...2. Table7 presents a summary of the above allocator algorithms, along with their speed, scalability, false sharing and blowup char- acteristics. As can be seen from the table, the algorithms closest to Hoard are Vee and Hsu, DYNIX, and LKmalloc.... ..."

"... In PAGE 11: ...2. Table7 presents a summary of the above allocator algorithms, along with their speed, scalability, false sharing and blowup char- acteristics. As can be seen from the table, the algorithms closest to Hoard are Vee and Hsu, DYNIX, and LKmalloc.... ..."

"... In PAGE 9: ... In four common biomedical image analysis tasks, we have illustrated that problems can often be approached by algorithms operating in either of these domains, with specific advantages and disadvantages. Table2 gives a brief summary of our examples and the respective COI and COD methods applied to them. 1 The accuracy of a volumetric inter-individual nonrigid transformation between two different subjects is not a particularly well-defined concept.... ..."

"... In PAGE 3: ... However, this comparison gives some interesting insights into the relation among the areas, and these insights are a good starting point for further and more complete comparisons. A summary of the comparison is given in Table1 . In our comparison, some concepts can be mapped easily onto concepts of other theories and systems.... ..."

"... In PAGE 3: ... Un- like Hoard, both of these allocators passively induce false sharing by allowing pieces of the same cache line to be recycled. Table1 presents a summary of allocator algorithms, along with their speed, scalability, false sharing and blowup characteristics. Hoard is the only one that solves all four problems.... ..."

"... In PAGE 8: ... 4. Tool bus integration with integrated front-ends Table1 summarizes the alternative tool interoperability solutions, in terms of the aspects of integration addressed and the level of flexibility of the solution. Table 1.... ..."

"... In PAGE 19: ... The latter approach searches for the subset of attributes over which naive Bayes has the best performance. The results, displayed in Figures 5 and 6 and in Table2 , show that TAN is quite competitive with both approaches and can lead to signi cant improvements in many cases. 4.... In PAGE 25: ... In particular, the cross- validation folds were the same for all the experiments on each dataset. Table2 displays the accuracies of the main classi cation approaches we discussed throughout the paper: NB: the naive Bayesian classi er BN: unrestricted Bayesian networks learned with the MDL score TANs: TAN networks learned according to Theorem 2, with smoothed parameters C+Ls: Chow and Liu method|Bayesian multinets learned according to Theo- rem 1|with smoothed parameters C4.5: the decision-tree classi er of (Quinlan, 1993) SNB: the selective naive Bayesian classi er, a wrapper-based feature selection ap- plied to naive Bayes, using the implementationof John, Kohavi, and P eger (1994) In the previous sections we discussed these results in some detail.... In PAGE 25: ... We now sum- marize the highlights. The results displayed in Table2 show that although unre- stricted Bayesian networks can often lead to signi cant improvement over the naive Bayesian classi er, they can also result in poor classi ers in the presence of multi- ple attributes. These results also show that both TAN as well as Chow apos;s and Liu apos;s classi ers are (1) roughly equivalent in terms of accuracy, (2) dominate the naive Bayesian classi er, and compare favorably with both C4.... ..."